Table 2.

Are there quantitative imaging biomarkers associated with the severity of neck and shoulder MSDs?

MSD Classification or diagnosis Author(s)	Biomarker	Severity measure mean & range if available	Symptom duration (mean & SD if available)	Major results (association between biomarker & disease)	Conclusion
Neck disorders and symptoms
Neck pain  Dibai Filho (2012) [37]	Skin temperature	NDI Cases: 8.33 (SD = 2.65) Controls: 2.27 (SD = 1.27)	Unknown	Correlation NDI and skin temperature in right versus left trapezius; thermal asymmetry: NS.	No
Neck pain  Elliott (2008) [46]	Fat index indicating fatty infiltration (relative fat)	NDI Cases: 21.9 (SD = 7.5) Controls: 45.5 (SD = 15.9)	Cases: 33.7 (20.6) mo Controls: 20.3 (9.6) mo	Within groups there was no association between NDI and fat levels; total upper fat, p = 0.15; total fat, p = 0.94. (No description of total vs upper fat in paper)	No
Neck pain  Javanshir (2011) [48]	Longus colli muscle CSA, APD, LD, and LD/APD	NDI 33 (SD = 0.5) VAS pain intensity 5.1 (SD = 0.8)	≥ 3 moa	NDI and CSA: rho = −0.45, p = 0.05, dominant side; rho = −0.48, p = 0.03, non-dominant side. NDI and APD: rho = −0.49, p = 0.03, dominant side; rho = −0.45, p = 0.05 non-dominant side. NDI and LD or shape ratio, NS. VAS and CSA, APD, LD or shape ratio, NS.	Yes
Shoulder disorders and symptoms
Rotator cuff tear (partial & full) or rotator cuff disease  Keener (2015) [34]	Rotator cuff tear enlargement (see paper for definition), shoulder pain	ASES score - American Shoulder & Elbow Cases: 98.3 (IRQ = 10) Controls: 100(IQR = 0) Surgeons SST score - simple shoulder test (score normalized to 100) Cases: 91.7 (IQR = 33) Controls: 100(IQR = 0) VAS pain intensityCases & controls: 1.0 (IQR = 0)	Unknown	ASES: smaller w. advancing tear type, p < 0.05; median decreased by 31.9 points w. new pain, p < 0.05. SST: smaller w. advancing tear type, p < 0.05; median decreased by 14.8 points w. new pain, p < 0.05. VAS: NS. w. advancing tear type; median increased by 3 points w. new pain, p < 0.05.	Yes
Shoulder impingement syndrome  Park (2007) [70]	Difference in mean skin temperature btwn shoulder sides	VAS pain intensity 6.6 (5.5–9)	22.6 (SD = 40.4) mo	NS differences btwn normal, hypothermic, and hyperthermic cases for VAS (all cases in this analysis, hypothermic defined as abnormally low temperature in involved side vs uninvolved side, hyperthermic defined as opposite of hypothermic).	No
Neck/shoulder disorders and symptoms
Neck/shoulder pain  Nilsen (2007) [42]	Finger skin blood flow	VAS pain intensity: maximal pain response, shoulder pain response, neck pain response Maximal pain response: Cases: 25 (SD = 20.0) Controls: 15 (SD = 16.1) Shoulder pain response: Cases: 17 (SD = 16.9) Controls: 10 (SD = 12.6) Neck pain response: Cases: 20 (SD = 20.3) Controls: 9 (SD = 11.1)	> 3 mo a	Maximal pain response: correlation w. finger skin blood flow response during first 10 min of the stressful task in cases, (rho = 0.52, p = 0.004), NS in controls (rho = 0.06, p = 0.71.	Yes
Neck/shoulder pain  Strøm (2009) [43]	Blood flow	Complaint severity score: neck pain, shoulder pain, musculoskeletal complaint severity index (MCI; mean of 7 pain items). 0 = no complaint; 12 = severe complaint. VAS during experiment: pain intensity and general tension Median (range) cases: neck pain 3 (1–12), shoulder pain 4 (0–12), MCI 2.5 (0.3–4.1) controls: neck pain 0 (0–6), shoulder pain 0 (0–6), MCI 0.5 (0–2.1) These scores are a composite measure of intensity x duration during the 4 weeks preceding the experiment.	3 subjects (2 women) reported having had shoulder and neck pain for less than 12 months, 13 (7 women) for 1–years, three (all men) for 5–10 years, and five (all women) for more than 10 years.	Cases: correlation between pain VAS and blood flow in active trapezius at end of work task (90 min): rho = 0.47, p = 0.03. No correlation at 15 or 45 min into work task, p > 0.05. Controls: no correlation at 15, 45, or 90 min into work task, p ≥ 0.05. Cases: correlation between pain VAS and blood flow in contralateral trapezius at end of work task (90 min): rho = −0.53, p < 0.01. No correlation at 15 or 45 min into work task, p > 0.05. Controls: similar direction of results, no rho or p-values supplied.	Yes
Neck/shoulder pain  Takiguchi (2010) [79]	Minimal & maximal standardized uptake values (SUV) of [18F]fluorodeoxyglucose (18F–FDG)	VAS	> 6 mo, with pain at least 1×/moa	Trapezius: mean SUVs & pain VAS (SUVmax: r = −0.603, p < 0.0001; SUVmin: r = −0.405, p < 0.0001). Gluteus maximus: mean SUVs & pain VAS, NS.	Yes

^a: inclusion criteria